A Student's Guide to Randomization Statistics for Multichannel Event-Related Potentials Using Ragu

• Main Authors: Marie Habermann, Dorothea Weusmann, Maria Stein, Thomas Koenig
• Format: Article
• Language: English
• Published: Frontiers Media S.A. 2018-06-01
• Series: Frontiers in Neuroscience
• Subjects: Ragu; randomization statistics; ERP; N400; EEG; microstates
• Online Access: https://www.frontiersin.org/article/10.3389/fnins.2018.00355/full

In this paper, we present a multivariate approach to analyze multi-channel event-related potential (ERP) data using randomization statistics1. The MATLAB-based open source toolbox Randomization Graphical User interface (Ragu) provides, among other methods, a test for topographic consistency, a topographic analysis of variance, t-mapping and microstate analyses. Up to two within-subject factors and one between-subject factor, each with an open number of levels, can be defined and analyzed in Ragu. Ragu analyses include all sensor signals and no a-priori models have to be applied during the analyses. Additionally, periods of significant effects can be controlled for multiple testing using global overall statistics over time. Here, we introduce the different alternatives to apply Ragu, based on a step by step analysis of an example study. This example study examined the neural activity in response to semantic unexpected sentence endings in exchange students at the beginning of their stay and after staying in a foreign-language country for 5 months.